Research reveals major flaws in research supporting Covid antibody testing

(CN)—While the world continues to tackle the Covid-19 pandemic, many praise the potential value of antibody testing, but a study published Wednesday shows that some tests are flawed Indicates that it should not be used.

Since the outbreak of the coronavirus earlier this year, few subjects have had more airtime and discussion as accessible and accurate Covid-19 test resources. Most of this focus has focused on molecular tests that determine whether someone is sick at the moment, but many serological tests that detect coronavirus antibodies in previously infected individuals. I was broken.

Serology is widely regarded as important to better understand how and when a population is most frequently exposed to the virus, and to help health professionals more accurately measure predisposition trends. It was British prime minister Boris Johnson, who suffered a seizure in Covid-19 and even stayed in intensive care for some time, personally praised these serological tests as “changing the game.”

But this optimism may not be backed by data.

Study published on Wednesday BMJ A team of international researchers has identified significant weaknesses in the rationale for serology. Evidence is particularly weak for point-of-care tests performed directly on patients outside of traditional laboratory settings, and researchers say the evidence does not even support their continued use.

The researchers made these messy observations by combining 40 different studies to determine the sensitivity or specificity of the Covid-19 antibody test compared to control tests. They found a study in medical databases and preprint servers from January to April.

They found that at least half of the research (mostly from China) was not peer reviewed. To make matters worse, most people showed a high likelihood or uncertainty of bias that would have falsely distorted the final result.

Of the 40 studies, the researchers note that only four outpatient-related data were included, and only two were tested at the time of care.

When they collected all sensitivity data from the study-a measure of people correctly identified as having the disease-researchers overlooked somewhere between 2.2% and 34% of patients with Covid-19 I have found that.

“Overall, the poor performance of Covid-19’s existing serological tests has led to medical problems, especially given the time and effort required for these tests and the challenging workloads facing many clinics. Raises questions about the usefulness of using such methods for formal decision making.”

The researchers also looked at the lateral flow immunoassay test, a point-of-care test that has recently been considered as a way to provide an “immune passport.” They were not impressed with the result.

Their data show that when using the lateral flow immunoassay test in a population with a 10% prevalence of Covid-19, 1,000 people tested that at least 31 people were immune, even if they were never infected with the virus. Will be declared. Researchers report that 34 people, who had another Covid-19, were actually infected, and have never been infected.

Therefore, governments around the world should place less emphasis on the accuracy of these tests, the researchers said.

“In addition, our findings should give a pause to governments considering the use of serological tests, especially point-of-care tests, to issue immunology “certificates” or “passports”. is. “

Fais Ahmad Khan, a researcher and scientist at the McGill University Health Center Research Institute, said that better clinical trials are a surefire way to better understand the efficacy of the trial. ..

“We need more rigorously designed clinical studies. What do these actually do when these serological tests are performed at the point of care and ultimately applied to the same population?” Will work,” Khan said in an email. “Such research can be carried out quickly with international cooperation. This is key to getting answers that are applicable to a variety of situations.”